Method and device for multi-user channel estimation

• US 6,765,969 B1
• Filed: 08/22/2000
• Issued: 07/20/2004
• Est. Priority Date: 09/01/1999
• Status: Expired due to Term

• Current Assignee: Motorola Solutions, Inc.
• Original Assignee: Motorola, Inc. (Motorola Solutions, Inc.)
• Inventors: Vook, Frederick W., Thomas, Timothy A.
• Primary Examiner(s): Bocure, Tesfaldet
• Assistant Examiner(s): TRAN, KHANH C
• Application Number:
  US09/643,541
• Time in Patent Office: 1,428 Days
• Field of Search: 375/148, 375/152, 375/259, 375/260, 375/267, 375/285, 375/346, 375/347, 375/349, 375/299, 375/295, 375/316, 375/340, 375/350, 370/320, 370/335, 370/342, 370/479, 455/63.1, 455/65, 455/67.3, 455/67.13, 455/501, 455/504, 455/506, 455/526, 455/553.1, 455/101, 455/103, 455/132, 455/134, 455/135, 455/226.1, 455/226.2
• US Class Current: 375/259
• CPC Class Codes: H04L 1/06 using space diversity
• CPC Class Codes: H04L 25/0204 of multiple channels
• CPC Class Codes: H04L 25/022 of frequency response
• CPC Class Codes: H04L 25/0224 using sounding signals

Abstract

The invention computes frequency-domain channel gains by compiling a set of estimated channel gains as a function of pilot sequences, a set of analytical channel gains variables, and a set of weighting coefficients variables. A plurality of weighting coefficients are computed as a function of time and frequency correlation functions, a noise correlation matrix, and pilot sequences. A weighting matrix is computed from the weighting coefficients. After receiving a training sequence from at least one transmitter, a received data matrix is computed from the training sequence. The weighting matrix and the received data matrix are used to compute the frequency-domain channel gains. The invention also provides a method for reducing the computational complexity of estimating the time and frequency response of at least one desired signal received by at least one antenna. Also, the time and frequency response of at least one desired signal received by at least one antenna can be both interpolated and predicted with the present invention.

Claims

1. A method of operating a communication system including at least one receiver and at least one transmitter comprising:

computing a plurality of weighting coefficients as a function of time and frequency correlation functions, a noise correlation matrix, and pilot sequences;

computing a weighting matrix from the weighting coefficients;

receiving a training sequence from at least one transmitter;

computing a received data matrix from the training sequence;

computing an estimated channel gain for each transmitter from the weighting matrix, and the received data matrix; and

passing each of the estimated channel gains to the receiver to recover a plurality of transmitted data symbols, wherein the estimated channel gains H_j(k_d,b_d) are computed as a function of Σ

Y(f_p,t_p,)q_j(d,f_p,t_p).

2. The method of claim 1 further comprising initializing a channel estimation device based on a pilot sequence of the transmitter, a desired set of time and frequency correlation functions, and a noise correlation matrix.

3. The method of claim 1 wherein the weighting coefficients are computed according to q_j(d)=Ψ

<sup>−

1</sup>α

_j(d).

4. The method of claim 1 wherein the estimated channel gains are computed according to Ĥ

_j(k_d,b_d)=YΨ

<sup>−

1</sup>α

_j(d) where Y is the received data.

5. A method of operating a communication system including at least one receiver and at least one transmitter comprising:

computing a frequency-interpolation matrix as a function of auxiliary frequency-domain channel gains variables, a set of time domain channel gains variables, and a predefined set of auxiliary weighting coefficients;

computing a final weighting matrix as a function of a provided primary weighting matrix, and the frequency-interpolation matrix;

receiving a plurality of training sequences from at least one antenna;

assembling a matrix of received training data from the training sequences;

computing a plurality of time-domain channel gains for each transmitter as a product of the final weighting matrix and the received data matrix;

computing a frequency-domain channel gains by taking a DFT of the time-domain channel gains; and

passing each frequency-domain channel gains to the receiver to recover a plurality of transmitted data symbols.

6. The method of claim 5 further comprising computing a plurality of time-domain channel gains as a function of the frequency-interpolation matrix and a plurality of provided channel estimates.

7. The method of claim 5 wherein the primary weighting matrix is E_j(k,b) and equals Ψ

<sup>−

1</sup>α

_j(k,b).

8. The method of claim 5 wherein the frequency-interpolation matrix is (G^HG)⁻

1G^H and is derived from $\stackrel{\min }{h_{j,m}<br><br>\left(b\right)}<br><br>{<br><br>H_{j,m}<br><br>\left(b\right)-{\mathrm{Gh}}_{j,m}<br><br>\left(b\right)<br><br>}^2,$

where h_j,m(b) is the set of time-domain channel gains, H_j,m(b) is the set of auxiliary frequency-domain channel gains, and G is the set of auxiliary weighting coefficients defined as $G=\left[\begin{array}{c}{f}^H<br><br>\left(0\right)\\ \mathrm{⋮<br><br>}\\ f^H<br><br>\left(K-1\right)\end{array}\right],f<br><br>\left(k\right)=\left[\begin{array}{c}1\\ {\mathrm{<br><br>}}^{\mathrm{j2\pi <br><br>}<br><br>\text{\hspace{1em}<br><br>}<br><br>k/N}\\ \mathrm{⋮<br><br>}\\ {\mathrm{<br><br>}}^{\mathrm{j2\pi <br><br>}<br><br>\text{\hspace{1em}<br><br>}<br><br>k<br><br>\left(L-1\right)/N}\end{array}\right].$

9. The method of claim 5 wherein the final weighting matrix Q_j(b) is Q_j(b)=A_j(b)G*(G^TG*)⁻

1, where A_j(b)=[E_j(1,b)| . . . |E_j(K,b)].

10. The method of claim 5 wherein the time-domain channel gains are h_j,m(b)=(G^HG)⁻

1G^HĤ

_j,m(b), where ${\hat{H}}_{j,m}<br><br>\left(b\right)=\left[\begin{array}{c}{\hat{H}}_{j,m}<br><br>\left(k_1,b\right)\\ \mathrm{⋮<br><br>}\\ {\hat{H}}_{j,m}<br><br>\left(k_K,b\right)\end{array}\right]$

and

Ĥ

_j,m(k,b) is the channel estimate on antenna m for desired transmitter j at frequency k and time b.

11. A method of operating a communication system including at least one receiver and at least one transmitter comprising:

computing a time-interpolation matrix as a function of auxiliary time-domain channel gains variables, a set of Doppler channel gains variables, and a predefined set of Doppler weighting coefficients;

computing a Doppler gain matrix as a function of a provided primary weighting matrix, and the time-interpolation matrix;

receiving training sequences from at least one antenna;

assembling a matrix of received training data from the training sequences;

computing a plurality of Doppler channel gains as a function of the Doppler gain matrix and the matrix of received training data;

computing a plurality of time-domain channel gains as a function of the Doppler channel gains;

computing a frequency-domain channel gains by taking a DFT of the time-domain channel gains; and

passing each frequency-domain channel gains to the receiver to recover a plurality of transmitted data symbols.

12. The method of claim 11 further comprising computing a plurality of Doppler channel gains as a function of the time-interpolation matrix and a plurality of provided channel estimates.

13. The method of claim 11 wherein the primary weighting matrix for desired transmitter j at frequency k and time b is Q_j(b) and equals Q_j(b)=A_j(b)G*(G^TG*)⁻

1.

14. The method of claim 11 wherein the time-interpolation matrix is (C^HC)⁻

1C^H and is derived from $\stackrel{\min }{g_{j,m}<br><br>\left(n\right)}<br><br>{<br><br>{\mathrm{Cg}}_{j,m}<br><br>\left(n\right)-h_{j,m}<br><br>\left(n\right)<br><br>}^2,$

where C (B×

B_T) and h_j(n) (B×

1) are $C=\left[\begin{array}{c}{c}^H<br><br>\left(b_1\right)\\ \mathrm{⋮<br><br>}\\ c^H<br><br>\left(b_B\right)\end{array}\right],h_{j,m}<br><br>\left(n\right)=\left[\begin{array}{c}{h}_{j,m}<br><br>\left(n,b_1\right)\\ \mathrm{⋮<br><br>}\\ h_{j,m}<br><br>\left(n,b_B\right)\end{array}\right],$

where h_j,m(n) is the auxiliary time-domain channel gains variables at a set of times b₁ through b_B, g_j,m(n) is a set of Doppler channel gains variables, and C is a set of Doppler weighting coefficients defined as $c<br><br>\left(b\right)=\left[\begin{array}{c}{\mathrm{<br><br>}}^{\mathrm{j2\pi <br><br>}<br><br>\text{\hspace{1em}<br><br>}<br><br>\mathrm{Vb}/N_k}\\ \mathrm{⋮<br><br>}\\ {\mathrm{<br><br>}}^{-\mathrm{j2\pi <br><br>}<br><br>\text{\hspace{1em}<br><br>}<br><br>\mathrm{Vb}/N_k}\end{array}\right],g_{j,m}<br><br>\left(n\right)=\left[\begin{array}{c}{g}_{j,m,-V}<br><br>\left(n\right)\\ \mathrm{⋮<br><br>}\\ g_{j,m,+V}<br><br>\left(n\right)\end{array}\right]$

and $C=\left[\begin{array}{c}{c}^H<br><br>\left(1\right)\\ \mathrm{⋮<br><br>}\\ c^H<br><br>\left(B\right)\end{array}\right],h_{j,m}<br><br>\left(n\right)=\left[\begin{array}{c}{h}_{j,m}<br><br>\left(n,1\right)\\ \mathrm{⋮<br><br>}\\ h_{j,m}<br><br>\left(n,B\right)\end{array}\right].$

15. The method of claim 11 wherein the Doppler gain matrix $Z_{j,l},\mathrm{is}$

 



Zj,l=Q~j,l

C*

(CT

C*)-1,

where

 



Q~j,l=[{Qj

(b1)}l



 



…



 





{Qj

(bB)}l].

16. The method of claim 11 wherein the Doppler channels $g_{j,v}$

(l)=[gj,l,v

(l)⋮

gj,M,v

(l)].

17. The method of claim 11 wherein the time-domain channel gains $h_{j,m}$

(n,b)

 



is

 



hj,m

(n,b)=∑

v=-V+V

gj,m,v

(n)



j2π



 



vb/Nk=cH

(b)

gj,m

(n).

18. A receiver for a wireless communication system comprising:

means for computing a plurality of weighting coefficients as a function of time and frequency correlation functions, a noise correlation matrix, and pilot sequences;

means for computing a weighting matrix from the weighting coefficients;

means for receiving a training sequence from at least one transmitter;

means for computing a received data matrix from the training sequence; and

means for computing from the weighting matrix, and the received data matrix estimated channel gains wherein the estimated channel gains H_j(k_d,b_d) are computed as a function of Σ

Y(f_p,t_p,)q_j(d,f_p,t_p).

19. A receiver for a wireless communication system comprising:

means for computing a frequency-interpolation matrix as a function of auxiliary frequency-domain channel gains variables, a set of time domain channel gains variables, and a predefined set of auxiliary weighting coefficients;

means for computing a final weighting matrix as a function of a provided primary weighting matrix, and the frequency-interpolation matrix;

means for receiving a plurality of training sequences from at least one antenna;

means for assembling a matrix of received training data from the training sequences;

means for computing a plurality of time-domain channel gains for each transmitter as a product of the final weighting matrix and the received data matrix; and

means for computing a frequency-domain channel gains by taking a DFT of the time-domain channel gains.

20. The means of claim 19 further comprising computing a plurality of time-domain channel gains as a function of the frequency-interpolation matrix and a plurality of provided channel estimates.

21. A receiver for a wireless communication system comprising:

means for computing a time-interpolation matrix as a function of auxiliary time-domain channel gains variables, a set of Doppler channel gains variables, and a predefined set of Doppler weighting coefficients;

means for computing a Doppler gain matrix as a function of a provided primary weighting matrix, and the time-interpolation matrix;

means for receiving training sequences from at least one antenna;

means for assembling a matrix of received training data from the training sequences;

means for computing a plurality of Doppler channel gains as a function of the Doppler gain matrix and the matrix of received training data;

means for computing a plurality of time-domain channel gains as a function of the Doppler channel gains; and

means for computing a frequency-domain channel gains by taking a DFT of the time-domain channel gains.

22. The means of claim 21 further comprising computing a plurality of Doppler channel gains as a function of the time-interpolation matrix and a plurality of provided channel estimates.

23. A computer readable medium storing a computer program comprising:

computer readable program code for computing a frequency-interpolation matrix as a function of auxiliary frequency-domain channel gains variables, a set of time domain channel gains variables, and a predefined set of auxiliary weighting coefficients;

computer readable program code for computing a final weighting matrix as a function of a provided primary weighting matrix, and the frequency-interpolation matrix;

computer readable program code for receiving a plurality of training sequences from at least one antenna;

computer readable program code for assembling a matrix of received training data from the training sequences;

computer readable program code for computing a plurality of time-domain channel gains for each transmitter as a product of the final weighting matrix and the received data matrix; and

computer readable program code for computing a frequency-domain channel gains by taking a DFT of the time-domain channel gains.

24. The medium of claim 23 further comprising of computer readable code for computing a plurality of time-domain channel gains as a function of the frequency-interpolation matrix and a plurality of provided channel estimates.

25. A computer readable medium storing a computer program comprising:

computer readable code for computing a time-interpolation matrix as a function of auxiliary time-domain channel gains variables, a set of Doppler channel gains variables, and a predefined set of Doppler weighting coefficients;

computer readable code for computing a Doppler gain matrix as a function of a provided primary weighting matrix, and the time-interpolation matrix;

computer readable code for receiving training sequences from at least one antenna;

computer readable code for assembling a matrix of received training data from the training sequences;

computer readable code for computing a plurality of Doppler channel gains as a function of the Doppler gain matrix and the matrix of received training data;

computer readable code for computing a plurality of time-domain channel gains as a function of the Doppler channel gains; and

computer readable code for computing a frequency-domain channel gains by taking a DFT of the time-domain channel gains.

26. The medium of claim 25 further comprising of computer readable code for computing a plurality of Doppler channel gains as a function of the time-interpolation matrix and a plurality of provided channel estimates.